Hacker News Reader

Comment by fermienrico | Hacker News Reader

People talk about the Metric system, but if we didn't have 10 fingers and if we went with a Base 12 system from the dawn of time, we would have something much better than the Metric system.

The duodecimal system is unique to the fact that the number 12 has more divisible factors than anything else that's manageable to count.

https://en.wikipedia.org/wiki/Duodecimal

Inches/Feet are based on Duodecimal system, but since we use Arabic numerals (base 10), it's obvious why inches/feet aren't the way to go and why Metric system wins.

There are still things that the rest of the world use Base 12 for - hours/minutes/seconds, 360 degrees, point font units, etc.

The reason why America hasn't adopted Metric system despite of it being the official system in the law is the same reason why the rest of the world hasn't gotten around to changing time/degrees into a decimal equivalent. It takes a tremendous amount of effort to turn the ship, the US being a giant ship to turn. America also leads in determining standards for Metric system around the world - NIST with collaboration from European counterparts work together to establish these standards. Universities in US, semiconductor industry, medical industry, etc. use metric units. Engineers in the US know and use the metric system.

If we didn't have Arabic numerals in Base 10, the American system is better :) *

But since we are all stuck with arabic numerals, the Metric system is better!

Edit: * I realized that its not that simple. Conceding this point, see responses. I was under the impression that most US units were consistent with base 12, but they're not - 1 lb = 16 oz (not 12) for example. A base 12 metric system would be ideal.

View on HN

If the American units were 'metric in base 12', there would be a lot more arguments for keeping them. But they really aren't - the units have no well-thought-out relationships to each other, they are just an amalgamation of different historical decisions, often taken in isolation. Inches to feet to furlongs to miles have one set of arbitrary conversions, while ounces to pounds to tons has entirely different conversions.

But contrast, a metric system in base 12 should still keep the niceties of the metric system - units of volume should correspond to units lengths of the side of a cube, units of weight should correspond to units of volume of water etc. And there should be a single rule for conversions, with standardized names for sub divisions.

These are the things that make the metric system superior, not its being in base 10. And these advantages far outweigh the advantage of being able to divide a sixth of a foot in a whole number of inches - particularly in the era where we carry arbitrary precision calculators in our pockets.

What's special in the rest of the world isn't just a metric system, it is one particular metric system: the SI system (also sometimes called the MKS system, for meter-kilogram-second). The SI system guarantees not only that we have nice multiples of powers of ten, but also that all scientific units are easily convertible to each other. This, for example, enables the fact that one ohm times one farad is exactly one second! That's why, for instance, you may have heard that NASA (or most any other scientific org in America) uses the so-called "metric system", again specifically meaning SI units. (There does in fact exist an arcane scientific system based on the foot-pound-second, but nobody - not even Americans - uses it.)

That is exactly what makes system so powerful. Ease of conversion between units and use in physics and engineering.

Only thing I hate about system is kilogram... The French... Couldn't do it properly. I really wish we would use grav instead...

>If we didn't have Arabic numerals in Base 10, the American system is better

While I'll agree that a base 12 system would be better, the American system would still be a bizarre hodgepodge.

Let's look at distance. First, a thou, or a thousandth of an inch. Breaking everything there. Then twelve inches to a foot, where things work. Then three feet to a yard? Twenty two yards to a chain, ten chains to a furlong and finally eight furlong to a mile.

And if you're wondering, yes I had to look up everything besides inch, foot, yard, and mile.

Haha yeah, same with temperature - Fahrenheit would still be a weird thing.

I dunno -- 0F is about as cold as 99% of humans will ever experince, and 100F is about as hot as most humans will ever experience. And a change in temperature of 1F is just about the minimum I can really detect.

With C, sure 0C is cold, but tons of people experience colder on a regular basis; and I've had situations where 20C feels a bit too cold, but 21C feels a bit too warm. Most thermostats denominated in C trigger on half a degree difference, not a full degree.

Celsius works better with one of the base materials in the world - water. At sea level 0 degrees water freezes and at 100 degrees it boils. For the same reason Celsius works better with metric system. "What feels cold" is not a good base point, because people are different and that feeling also depends on humidity and wind.

The GP didn't say any of that; they simply said Fahrenheit was "weird". I disagree; Fahrenheit simply human-centric, and once you get used to it, is a lot more practical for day-to-day human experience than celsius.

You're pointing out how useful it is when doing science with water. Sure. And since science is important, and it doesn't make much sense to keep two systems around, then yeah, let's go with celsius as the standard. But let's not lose sight fo the fact that we actually are giving up a more human-focused system when we gain this scientifically-focused system.

Speaking from the human experience, "0 degrees Celsius means the road could be icy" is more useful than "68 degrees Fahrenheit is normal room temperature", wouldn't you agree?

"The road could be icy" is certainly a useful data point. But "70% of the way from 'wow this is cold' to 'wow this is hot' is a moderately comfortable temperature" seems reasonable to me.

(68 is a comfortable temperature for men wearing suits... a better more general temperature is probably 72.)

This all puts way too much stock in temperature being what decides if we go "wow this is cold/hot." Wind, humidity, and other assorted factors play too large of a role to value basing our temperature system on it.

Zero F with no wind can feel like a nice day compared to zero C with tons of wind.

Even Americans use decimal based systems in engineering. A lot of measurements are defined in thousands of an inch. Why didn't they go with a "superior" system with more divisible factors?

The reality is that metric is a standardizing measurement. There is one base unit, changes in scale merely involve a factor denoted by a prefix, it doesn't try to be smarter than absolutely necessary and just sticks to decimal.

I agree that 10 as the base is suboptimal, however everyone using the same base has great network effects. That’s what the other 190-ish countries are annoyed about

The thing is, the US tried to adopt the Metric system but with number of failures: https://en.wikipedia.org/wiki/Metrication_in_the_United_Stat...

Smaller nations have no problem changing everything. The marginal benefit is too small for a giant to turn on a dime.

A lot of things in US are already both ANSI/Metric:

- Electronics and PCB manufacturing (accepts both, but its a bit messy). Headers are 0.1" but BGA pitch is 0.4mm for example.

- Hardware stores have both standard and metric sizes for fasteners

- Liquids (both oz/ml has to be printed by law)

- Car speed gauges (both mph/kph printed by law)

- Food items are all metric. A Coke can has 42 grams of sugar. There is no oz printed in nutrition labels.

It's kind of a hybrid system.

However, there are exceptions for the follow which are still stuck:

- Temperature, weather is always in Fahrenheit. Even in recipes.

- Paper sizes are not adopted to ISO standards.

- Pressure is almost always in psi

> Smaller nations have no problem changing everything.

And some smaller nations (the UK, for example) have never changed "everything" and likely never will.

In about 4 months the UK is scheduled to leave the EU.

Probably some portion hope to revert completely when they do.

Probably some portion hope that they rejoin the EU with another vote; and in the process get dragged the rest of the way to the engineering age measurements in the process.

This isn't about Brexit or politics. For example, Japan hasn't fully converted to metric, either:

https://resources.realestate.co.jp/buy/guide-to-japanese-apa...

> The size of a Japanese-style room is measured by the number of tatami mats:

>> Four mat room = 6.12 square meters = 66 square feet

>> Six mat room = 9.18 square meters = 98.8 square feet

> Most Japanese-style rooms in modern apartments are four or six mats in size.

Well, for rooms with actual tatami mats, you are not really interested in the area, but rather how many mats will fit. Thus, a 24m^2 (12 mat) room with the wrong aspect ratio will not fit 12 mats. (I'm using the 1 mat = 2m^2 conversion for convenience and I actually use 2m^2 mats in my tatami room.) Practically, mat sizes can vary a bit and can be custom sized. [1]

Even though today only a minority of apartment rooms are actually used with tatami it is still customary to give all rooms in mat units and usually in m^2 too.

BTW that's similar to how non SI units are still used in other countries nowadays, e.g. (metric) pounds and half-pounds are still used in context of groceries in colloquial German.

[1] https://en.wikipedia.org/wiki/Tatami#size

The UK left the EU at the end of January 2020. The EU is currently treating us as having various member agreements in place until the end of December 2020 but we have already left at this point.

>Probably some portion hope to revert completely when they do.

That would make trade deals with the EU impossible.

So, China or India don't use the metric system?

One interesting fact about US food labeling requirements is that manufacturers are legally required to use incorrect conversion ratios.

Per the law 21 CFR 101.9(b)(5)(viii) [0]:

> (viii) For nutrition labeling purposes, a teaspoon means 5 milliliters (mL), a tablespoon means 15 mL, a cup means 240 mL, 1 fl oz means 30 mL, and 1 oz in weight means 28 g.

In reality:

* 1 fl. oz. ≈ 29.57 mL [1] (the volume measurements all build off of one another so I won't bother converting more than one)

* 1 oz. ≈ 28.35 g

In other words, volume and mass measurements are legally required to be off by ~1.4% and ~1.3% respectively.

[0]: https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/cfr...

[1]: https://en.wikipedia.org/wiki/Fluid_ounce

> Smaller nations have no problem changing everything. The marginal benefit is too small for a giant to turn on a dime.

I don't think it is the size of the US that is the problem, I think the real problem is the US political culture.

The most successful approach to metrication is (a) do it in a big-bang, all at once, not gradually (b) make it apply uniformly nation-wide (c) make it apply across as many industries/sectors as possible, with as few exceptions as possible (d) take a predominantly mandatory rather than voluntary approach to metrication. If the US had followed that formula, it would very likely have succeeded, but it never had the political will to go down that path. Instead it opted for a piecemeal, voluntary approach, which was never likely to be a great success

In a parliamentary system such as the UK, Canada, Australia, etc, cabinet decides on a policy, and then the government usually has the numbers in the legislature to enact whatever policy cabinet decides. By contrast, in the US, the executive (President and cabinet) and the legislature (Congress) are two arms-length branches of government, with far less coordination, and it becomes much harder to push policies through unless strong interests are backing them.

Hence, in the US, it is basically impossible to get any controversial policy implemented unless either it has corporate lobbying behind it, or else it is some kind of emotional issue which attracts the attention of activist groups. Corporate lobbyists don't support mandatory metrication (it will cost business money, at least in the short term); metrication doesn't induce the kinds of emotions that issues like abortion or racism do, so there is little activist energy pushing it.

Also, the US historically has a much more hostile attitude to technocratic bureaucracy than most other Western countries do. In other countries, the scientific/technological elite consensus is "this way is better", and the government endorses it, and so the general public falls into line behind the government. In the US, there has always been much more resistance among the general public to those kind of elite proposals. (That said, I think in more recent years the rest of the English-speaking world is becoming more like the US in this regard; I think the Internet has helped export that US culture of distrusting technocratic elites to the rest of the English-speaking world.)

> Smaller nations have no problem changing everything. The marginal benefit is too small for a giant to turn on a dime.

Every other nation seemed to manage it. It's just the usual American exceptionalism. There's nothing special about the USA that makes it harder to implement. There's just just a stubborn belief that there is, which becomes a self-fulfilling prophesy. A bit like all the other policy issues which the rest of the world has solved but the US still argues about.

I really don’t think that’s the issue. Even in the US, people complain constantly about the units.

The stubbornness you speak of is actually more to do with not willing to accept the costs of the change. World’s largest economy can’t just change it all at once even though majority of the people support it.

No one in the US argues that the Metric system isn’t superior. It doesn’t take a genius to figure out that multiples of 10 in a base 10 system of numbers makes everything easy.

In fact, the superiority complex comes from non-US folks thinking “look at those dumb Americans and their stupid units”.

Certain people think “Americans just don’t get it”.

Are you content that US and majority of the countries drive on the same side? https://en.m.wikipedia.org/wiki/Left-_and_right-hand_traffic

Left vs right hand drive is a poor comparison. The choice is entirely arbitrary, and nobody can seriously say that one is "better".

I am talking about adoption and standardization, not whether one is better than the other. If right hand side was “better”, it would be pretty much impossible for US/EU to change the driving side. Can you imagine the costs? Even if it was better.

Yes, but my point is that it isn't an example of the US adopting a standard. It's the US just happening to have chosen the more popular of two arbitrary options.

And the longer you wait, the harder it gets. They did at least start on the road to metric so perhaps in a generation or two, we are there!

What's a concrete eli5 example of how a base 12 system would be a benefit? The number 12 still exists. The base just dictates when the order of magnitude changes, so larger numbers can be represented with slightly fewer digits, but I don't see much more. Am I missing something?

If the numeric system was 0123456789AB and 10 (would be 12 in decimal), then 10 would have 2, 3, 4, 6 whole number fractions.

10/2 = 6

10/3 = 4

10/4 = 3

10/6 = 2

That's two more whole number fractions than 10 (decimal) which is only divisible by 2 and 5.

Meanwhile, the benefit of a system like metric is decimals rather than fractions. It is trivial to change the resolution of a measure or number.

E.G.

The speed of light is 299,792,458 meters per second.

The speed of light is 299,792.458 KM per second.

The speed of light is about 300 M meters per second.

That’s not true. You can have decimals for any base.

10/5 in Base 12 is 2.49724972497...

Or binary (Base 2)

1011/101 = 10.0011001100...

Duodecimal metric system would have the same features that you’re talking about (which have nothing to do with the base).

The metric system reminds of engineers doing UX.

Imperial system reminds of UX people doing engineering. With one system you actually get a working program, guess which one...

metric

Metric makes sense in science, sure. But our system makes more sense to humans. A gallon of gas is easier than 3.8 liters. A pint of beer is easier than its metric equivalent. As a gearhead, I absolutely hate the metric/standard difference. But if metric wants to take over the world, I feel they need more human submeasurements.

That doesn't make sense. A liter is better to imagine than a gallon (which is a huge jug). A pint of beer is same as saying 500ml of beer - both are equally easy to grasp.

Beer pouring is inexact. 500 ML is never 500ML. A system based on halfs is inexact but more accurate.

Have half a litre then. Speaking as someone from the only country that legally mandates beer in both pints (on draught) and metric (in bottles), I prefer a (UK, 20oz) pint if I'm drinking, because it's larger and I'm thirsty. For every other reason though litres are easier to reason about. The US 16 oz pint doesn't even have the benefit of being larger!

But usually in British pubs, beer is served in glasses that are a pint up to the brim, so if you are served beer with any head at all it’s probably closer to half a litre than a pint.

It's physically impossible to measure exactly 500.(0) mL [500.0000000000… mL with infinite number of zeroes] of any liquid with absolute accuracy. It's just the law of metrology. There always will be an error, you will always be able to create better tools to add extra decimal places to your measurement and find more error. The question is how much of error you can tolerate in any given situation.

The error has nothing to do with units of measurements chosen. You cannot measure exactly 1.(0) gallon [1.0000000000… gallon with infinite number of zeroes] of liquid with absolute accuracy either.

In engineering this would be resolved with something like 500 ± 10 mL.

In consumer goods there are either regulations or you get, say, 505 mL so you have nothing to complain about. And you would never be able to tell the difference.

By the way, ML is megaliter. Milliliter is mL.

That's a lot of beer.

524'151 tuns? With sigfigs, call it 500 kilotuns[1].

[1] unfortunately https://en.wikipedia.org/wiki/Tun_(unit) leaves me no wiser as to what volume the unit "shit tun" might measure.

The London beer flood was only about 1 Ml https://en.wikipedia.org/wiki/London_Beer_Flood

All this is saying is "I'm more familiar with a gallon than a liter." There's nothing that makes a gallon more sensible to humans.

It does if you have 8 pints, and pints are far more convenient as a size for many liquids than a litre.

Why are pints more convenient?

It's easier to carry beer and milk in those sizes and they're about the right amount for use. A cup of beer is too small, a litre too big as you'll end up with a flat beer by the end. A pint of milk will give 2 people their breakfast and enough left over to add for tea for the day. Rare to find some off milk in the bottom of a pint bottle.

I'm guessing that why they were sold in those amounts vs some other choice. It's just my experience though.

> It's easier to carry beer and milk in those sizes and they're about the right amount for use.

The US pint is 16 ounces, vs the Imperial pint of 20 ounces. That's because no one really wants that much American beer.

Yet the Germans get along just fine with steins of half litre or full litre. Maybe they're doing beer wrong.

In Australia, standard beer glass size varies from state to state -- here in NSW we have a schooner (at 425ml) or if you're after something smaller, a middy (285ml).

Using metric does not mean you need to always and only drink 1 litre of beer, drive at 100km an hour, constantly stoop so you're 1m tall, etc.

> Yet the Germans get along just fine with steins of half litre or full litre.

Good for them.

> In Australia, standard beer glass size varies from state to state -- here in NSW we have a schooner (at 425ml) or if you're after something smaller, a middy (285ml).

I know. You can buy a yard of ale in the UK at participating outlets. None of that has any relevance to whether a pint is more convenient than a litre. A schooner is more convenient than a litre.

> Using metric does not mean you need to always and only drink 1 litre of beer, drive at 100km an hour, constantly stoop so you're 1m tall, etc.

I'm arguing against dogmatism, not for it.

> None of that has any relevance to whether a pint is more convenient than a litre.

But one pint versus one litre doesn't have relevance to anything, does it?

It seems you've heard the suggestion to adopt a sensible, integrated, and (much more) widely used measurement system -- and are choosing to argue against specific numbers / quantities of things.

Originally you said 'a litre [of beer] is too big as you'll end up with a flat beer'.

Metric doesn't mean you must sell, buy, or drink beer in round numbers.

It's not just about the gallon, but its base 2 subunits.

Where I live a pint is 0.5L and a metric cup is 0.25L.

Any argument that is "makes more sense to humans" that is actually based on repeated exposure to the concept/metric since young age fall flat on itself.

It makes more sense to YOU because you've used it since early life, so you have created an mental model for it. Countries with metric system, people have done the same.

As one example, In a recent after-work conversation, some one mentioned the stupidity of the "stone" measurement, upon the Italian, Canadian, Japanese, American, Swedish, Mexican, Chinese and the Russian all agreed it was the most non-intuitive stupid thing they have ever heard. Except the Brit at the place, who (in vain) tried to explain how it made sense and made "more sense to humans"

You're spot on with the mental models. If I try to imagine a gallon, I think of 2x2L bottles, one of them not quite full. A pint and I imagine a 600mL soft drink or water bottle. When you start having to do double conversions (eg miles per gallon) it becomes hopeless - that means nothing to me - and I have to convert to l/100km.

These "intuitiveness" arguments are silly. It's whatever you're used to. So with that realisation, we may as well choose the more universally self-consistent and useful system, which is indisputably metric.

> A pint and I imagine a 600mL soft drink

Careful, that's a British pint. An American pint is closer to 400mL.

> It makes more sense to YOU because you've used it since early life, so you have created an mental model for it. Countries with metric system, people have done the same.

There was an xkcd comic with lists of ready-to-go mental models about metric units for people who grew up without them:

https://xkcd.com/526/

I'm not British, but I do agree that the stone measurement does make sense in a human way. It's pretty easy to see how it came about. Oh take a rock about the size of x(whatever x was agreed upon at the time), and see how many that item weighs. In a time without precise machinery, or any machinery at all, you can see how such a system was thought up. Do you disagree with the above, or do you just think in the modern world it shouldn't exist anymore?

>. Oh take a rock about the size of x(whatever x was agreed upon at the time), and see how many that item weighs.

I do not disagree with the above, I can totally see that being done (hell, when we didn't have a measure out while hiking with my friends we used "5 iPhones long" as a temporary measurement)

Problem is of course, that I can't see how that measurement can ever scale well beyond one medieval towns market place as a measurement, let alone a country, kingdom or a territory without confusion. It is ambiguous. Just as the example you said, it is easy for a fraudster to pick up any rock and claim that one to resemble the standard rock unit and get some unfair trading advantage. It is too easy to trample with. The purpose of measurement is exactly to REMOVE ambiguity, and establish a STANDARDS just as we established common languages for communication.

And on a side note, usage of the "stone measurement" was officially prohibited for commercial use in 1985 in UK. It is apparently still used for body-weight, but I guess there ambiguity can be preferred ;)

Updating measurement unit according to societal needs is only natural, and the purpose of the Stone story in my comment was that, even among so many different national (with different culture, language, metric system) they thought the Stone measurement made no sense today. Except the Brit that had a mental model for it. Which was the point: sometime mental model can seem like an good argument, but as the Canadian and American also thought stone was dumb, (While everyone else in the room thought imperial units were dumb) they kind of proved that their argument for imperial units was also not found on good basis.

I could just as well say that 1 liter is easy, but 0.264172 gallons is complex.

Using a conversion as a justification just isn't a fair comparison.

I'm pretty sure metric took over the world a long time ago?

Why is a gallon easier than a liter?

One desirable property with the SI system is how compatible they are: a cube with 10cm sides is a liter, so it’s easy to imagine the space requirements for a 40 liter tank. A gallon is 231 cubic inches, which is much less natural to work with.

Because a gallon is divisible in human unit. A quart is a quarter or a gallon, a pint is half that. 250ml is both exotic sounding and guaranteed not as exact when measured in halves.

>Because a gallon is divisible in human unit.

And the litre isn't? Sorry, I fail to see your argument here.

And why would a quarter of a gallon be more convenient or 'divisible in human unit' than a quarter of a liter...?

Say, I walk into a bar. I order a beer. Here (in The Netherlands), since I didn't specify, I get a small beer of whatever's on tap. A small beer is 250ml-350ml, depending on the establishment. I could also order a half liter (500ml) or ask for a pitcher. So, just as in the US you don't order "can I have 128 fluid ounces of beer" you don't order "250ml of beer" in the rest of the world. Also, the fact that it 'sounds exotic' says more about you than about the system... and 'guaranteed not as exact'? Really? Let's say you have to measure a gallon of water, and I have to measure a liter. I just whip out a digital kitchen scale and measure 1000 grams. Which is EXACTLY the same amount as 1000 milliliters. Cause, you know, metric units are interchangeable. I double dare you to come up with a more exact measurement in Imperial units using simple kitchen tools.

> In The Netherlands [...] a small beer of whatever's on tap [...] is 250ml-350ml

Traveler beware: Just because it's true that a Dutch "vaasje" glass is 230-330ml[1] the actual volume of beer you get is at best 80% of that. The Dutch pouring practice of handing patrons a glass of beer with up to 1/4 receding foam on the top would cause a riot across the channel.

> [...]and measure 1000 grams. Which is EXACTLY the same amount as 1000 milliliters[...]

No, not exactly. Not even under standard conditions. It's one of the unfortunate warts of the modern metric system that the actual value drifted from exactly 1000, almost, but not exactly[2].

1. https://nl.wikipedia.org/wiki/Vaasje

2. https://en.wikipedia.org/wiki/Density#Water

I'll assume that you are not being difficult for some malicious reason and answer you as simply as possible:

You order a quart when you want a quarter of a gallon in the US.

You order a "Viertelliter" (quarter of a Liter" when you want a quarter of a Liter in Germany.

How is one quarter less precise / more exotic than the other?

So, I'm only going to try to explain my view once more to avoid a million downvotes replying to all, but I'm not intentionally being difficult.

Most places, I think, most people don't talk in precise units at all. Give me a glass of water. Put in a pinch of salt. One scoop of coffee. Handful of peanuts. It is my personal experience that this tracks pretty well across any culture. In my opinion, having these oddball measuring units is an easier mental model. A couple tablespoons is human sounding. 30ml is not, even if you have a 15ml spoon. A cup of water sounds human, a quarterliter does not. I'm not trying to insult you or your system, I'm just expressing my own opinion, which as pointed out, probably comes from my own education and experience.

Surely you jest.

You've never left the US, have you? No one uses 3.8 liters of gas, they obviously use 1 liter of gas, or 10 liters of gas. Most cars have about 40 or 50 liter tanks, and gas is paid by the liter.

Juice is also bought by the liter, usually in 2L bottles for a family dinner, or 500mL bottles for one person alone.

I've been all over the world. Don't make assumptions based on your own argument.

Well you sort of betrayed yourself imagining all those hapless foreigners, struggling to fill their cars in multiples of 3.8L ..

In most countries you could go your whole life without having to know what a "gallon" is or even hearing the word.

And have you seen anywhere people using 3.8l or 0.334 meters or 12.5cm as common measurements? Or in general, people having problems with expressing quantities in simple multiples or fractions of liters, (kilo)grams, and meters?

I do understand that it is hard to adjust intuitions from one system to the other, but I can't imagine how someone can think that people have problems using the measurement system they learned in primary school.

2l is very large in my (Dutch) idea. Orange juice in my country is typically sold in 1l cartons and more recently also 1.5l. But the latter only for the most popular flavors. The same is true for milk.

I think I should have soda 'soda', not 'juice'. Here in Romania, things like orange juice and milk are also usually sold in 1l bottles. It's things like Coca Cola and Fanta that are typically sold in either 0.5l or 2l bottles (even 2.5l around holidays; but 1.5l is also becoming increasingly common).

It's common in the US and Australia and, I suspect, any large producer of fruit with a normal culture of drinking juice. Anywhere that imports 1l seems to be more the norm.

Strange what you say about milk though - I've usually seen the 2l form factor anywhere westerners are common. Could be a local convention?

The EU had rules about standard sizes for most packaged food, which probably influenced this. I can't find what they were for fruit juice -- the rule was abolished in 2007 (or 2002?), since surveys showed people who cared about the price would look at the "per 100mL" indication on the shelf anyway.

Different European countries produced different fruits (citrus, apples etc), but I've not noticed that e.g. apple juice is in larger containers in Britain, and orange juice in Spain. If anything, the trend is to smaller containers, for variety and health.

https://ec.europa.eu/growth/single-market/goods/building-blo...

In Germany you have 1l packs for milk and 0.5l packs for chocolate milk...

The US system's main strength is that there are a bunch of human-scale subunits (gallon, pint, cup, etc.). However, the issue is that there's a lot of those subunits (some useful, some not).

Converting between the units is a pain: not because the conversion is hard (the conversion ratios are usually whole numbers), but rather because there's a lot of units and it's difficult to remember how they relate to each other (e.g. what's the relation between a pint and a gallon?).

Why is the gallon a "human sized unit"? What does it even represent? A hat size?

And it's weird to call a unit "human sized" when there are different versions of it.

A US gallon is 3.785 litres. A UK gallon is 4.546 litres.

Which if these is more human-sized?

>Which if these is more human-sized?

The US gallon since they've got more people per capita.

It's a useful quantity because the variety of units allow for the speaker to drop the numerical portion of a quantity of a substance.

Meaning, the phrase "gallon of milk" is ever so slightly easier to say/verbalize than "four liters of milk." Or alternatively, "half gallon of milk" vs "two liters of milk."

Likewise: "pint of beer" vs. "<half liter/5 deciliters> of beer", "1 cup of flour" vs. "<quarter liter/250 mL/25 centiliters/120 grams> of flour".

But you're not saying "four litres of milk" that often. You're saying "a carton of milk" and typically the people understand that that is a litre or 1.5l or whatever it is you usually buy. As for beer, again, where I drink it's typically either unspecified or a "small", being 25cl and 15cl respectively (this being a nerdy craft beer bar where smaller sizes are preferred.) If you do want to say the size, you can just say 25 or 15, no one is going to think you want 15 litres because that'd be weird.

> A gallon of gas is easier than 3.8 liters

Most people in the world just buy their petrol per liter directly and it's priced at <currency amount> / liter, there is no conversion needed, the "gallon" is just an arbitrary measure.

And we do not measure fuel consumption in "miles per gallon" , but "liters per 100km".

>And we do not measure fuel consumption in "miles per gallon" , but "liters per 100km".

Actually that is the "official" way of measurement (what you see on the car specifications or what the on board computer tells you), in speaking it is more common to say that you made 22 Km/liter (or 18, etc.) than to say you used 4.55 liters/100 Km (or 5.56 liters/100 Km).

I've never heard anyone say that. Who are these people that are reading fuel consumption in their dashboard and doing the mental arithmetic to convert that to km/liter? I've only heard "my car spends about 5-8", meaning 5-8 l/100 km in casual conversation. I.e. the number you'd read directly from the dashboard.

Well, maybe it is just local, but before you had liters per 100 km or actually a dashboard at all (not so many year ago) all you had was a lousy analogic something with a hand that went from F to E and the only way to calculate (approximately) fuel consumption was to know how many liters you put on the refill and have the km since last refill.

So you had (say) 30 liters needed to refill and travelled for 450 Km, now is it easier to divide 30 by 450 and multiply by 100 ( or divide 30 by 4.5) or divide 450 by 30?

Besides the difference between (say) 18 and 19 Km/l is (IMHO) easier to visualize/appreciate than the difference between 5.56 and 5.26 l/100 km.

>Most people in the world just buy their petrol per liter directly ...

It depends.

Here (Italy) most petrol stations are self-service/with automated machine for payments.

If I buy in cash I buy gasoline "per money", i.e. 20 Euro or 50 Euro, you don't have any alternatives if not multiples of the bills.

If I buy paying with a card then I still try (roughly) to meet the amount.

I am not really aware that today 50 Euro will give me (say) 27.476 liters while last week (still say) it gave me 27.187 liters.

> A gallon of gas is easier than 3.8 liters

I'm pretty sure this is an attempt at humor. Or I'm not sure but I hope.

> As a gearhead, I absolutely hate the metric/standard difference.

To take the example of the F150 there's a 26 gallon fuel tank, but then the engine size in in liters, transmission fluid is in quarts, windshield washer reservoir in liters etc.

Even on the most popular American-made car in the US you need to deal with an absolute mess of different measurement systems, on e.g. a German-made car all of these would be metric.

This is still better than tires, where the diameter seems to be listed in inches around the world even though other measurements such as the width are usually in millimeters everywhere.

Tires are their own special measurement hell. A 17 inch tire does not have a diameter of 17 inches. Those 17 inches are the diameter of the rim it's supposed to sit on[1].

Thus e.g. a 245/45 R17 tire has a diameter of around 26 inches. As you can deduct from the "simple" formula of:

    245/10*45/100/2.54*2+17 =~ 25.7

I.e. the width is 245mm, the cross section of the tire is 45% of that, or ~110mm, multiply that by 2 and add the 17 inches and you've got the outer tire diameter.

1. https://www.moderntiredealer.com/8644/understanding-tire-dim...

A gallon of gas is 3.785 L in the US. In the rest of the British Empire it was 4.546 L. It definitely was not easier back when both gallons were in common use.

And by "pint of beer", do you mean an Imperial pint (568 mL) or a US liquid pint (473 mL)?

Reply on news.ycombinator.com